Delayed production can substantially alter the qualitative behaviour of feedback systems. Motivated by stochastic mechanisms in gene expression, we consider a protein molecule which is produced in randomly timed bursts, requires an exponentially distributed time to activate and then partakes in positive regulation of its burst frequency. Asymptotically analysing the underlying master equation in the large-delay regime, we provide tractable approximations to time-dependent probability distributions of molecular copy numbers. Importantly, the presented analysis demonstrates that positive feedback systems with large production delays can constitute a stable toggle switch even if they operate with low copy numbers of active molecules.

The chemical master equation is a fundamental equation in chemical kinetics. It underliesthe classical reaction-rate equations and takes stochastic effects into account. In thispaper we give a simple argument showing that the solutions of a large class of chemicalmaster equations are bounded in weighted ℓ1-spaces and possess high-ordermoments. This class includes all equations in which no reactions between two or morealready present molecules and further external reactants occur that add mass to thesystem. As an illustration for the implications of this kind of regularity, we analyze theeffect of truncating the state space. This leads to an error analysis for the finite stateprojections of the chemical master equation, an approximation that forms the basis of manynumerical methods.

In this paper we study the error rate of RNA synthesis in the look-ahead model for therandom walk of RNA polymerase along DNA during transcription. The model’s centralassumption is the existence of a window of activity in whichribonucleoside triphosphates (rNTPs) bind reversibly to the template DNA strand beforebeing hydrolyzed and linked covalently to the nascent RNA chain. An unknown, butimportant, integer parameter of this model is the window size w. Here, weuse mathematical analysis and computer simulation to study the rate at whichtranscriptional errors occur as a function of w. We find dramaticreduction in the error rate of transcription as w increases, especiallyfor small values of w. The error reduction method provided by look-aheadoccurs before hydrolysis and covalent linkage of rNTP to the nascent RNAchain, and is therefore distinct from error correction mechanisms that have previouslybeen considered.